In the Literature

The Gut–Vascular Barrier

Spadoni I, Zagato E, Bertocchi A, et al. A gut-vascular barrier controls the systemic dissemination of bacteria. Science 2015; 350:830–4.

The gastrointestinal mucosa, particularly that of the lower gastrointestinal tract, is continuously exposed to enormous numbers of microorganisms against which it must provide an effective barrier to their translocation into the portal venous drainage. Spadoni and colleagues have now demonstrated the presence of a gut–vascular barrier of mice and men that prevents the translocation of bacteria and of bacterial products.

Using a murine model, the investigators demonstrated that this barrier prevented translocation of molecules of approximately ≥70 daltons. The barrier, like other endothelial barriers such as the blood–brain barrier, consists of a tight junction and an adherence junction, that act as functional molecular assemblies that control paracellular passage of fluids and solutes. Studies of human colonic cell cultures demonstrated similar assemblies, indicating the presence of a gut–vascular barrier in humans as well. This barrier was disrupted by Salmonella typhimurium in both experimental systems.

Spadoni and colleagues examined patients with celiac disease on a gluten-free diet but who had elevated serum concentrations of alanine aminotransferase, and found that they had evidence of impairment of their endothelial barrier function. This finding suggests the possibility that disruption of the gut–vascular barrier may be responsible for hepatic dysfunction in other settings as well, such as sepsis, a circumstance in which the injury may progress to hepatic failure. Translocation of bacterial products (and sometimes intact bacteria) has been demonstrated to contribute to complications of end-stage liver disease such as encephalopathy and hepatorenal syndrome.

The identification of the gut–vascular barrier opens the door to further investigation including approaches to its manipulation. If, for example, it plays a role in the hepatic dysfunction seen in sepsis, enhancement of its function may prove therapeutically important. The role of the intestinal microbiome in preservation of barrier function also deserves investigation.

Cellulitis or Pseudocellulitis?

Strazzula L, Cotliar J, Fox LP, et al. Inpatient dermatology consultation aids diagnosis of cellulitis among hospitalized patients: a multi-institutional analysis. J Am Acad Dermatol 2015; 73:70–5.

Strazzula and colleagues retrospectively examined the records of all 74 inpatients at 4 US tertiary care centers for whom a dermatology consultation for the evaluation of cellulitis was requested in 2008. These represented 5.2% of all 1430 inpatient dermatology consultations. The consultants determined that only 25.7% of the 74, in fact, had cellulitis, a proportion that ranged among the institutions from 6.7% to 44.4%. A comparison of the patients with and without cellulitis (pseudocellulitis) revealed no significant differences, including in the presence of leukocytosis, which was detected in only a small minority of both groups (5.3% and 16.4%, respectively; P = .44). The most frequent causes of pseudocellulitis were stasis dermatitis (30.9%), contact dermatitis (14.6%), and inflammatory tinea (9.1%). Additional diagnoses included drug eruption, erythema chronicum migrans, psoriasis, vasculitis, and lymphedema.

The frequency of misdiagnosis reported by Strazzula and colleagues is remarkable, but the sample is not unbiased. Presumably only patients for whom the diagnosis seemed uncertain triggered a request for dermatology consultation. A previous study by David et al [1] evaluated all 145 patients admitted from the emergency department of 2 hospitals with a diagnosis of cellulitis and found, after evaluation within 24 hours by either a dermatologist or infectious disease specialist, that the diagnosis was incorrect in 41 (28%). Among the 41 in that series with pseudocellulitis, the most frequent causes were stasis dermatitis in 15 (36%), followed by deep venous thrombosis, nonspecific dermatitis, thrombophlebitis, and trauma-related inflammation—each in 5%. All 4 patients with bilateral cellulitis were diagnosed by the specialist as having stasis dermatitis.

Unsurprisingly, the misdiagnosis of cellulitis also occurs in the outpatient setting. Arakaki and colleagues [2] randomized patients given a diagnosis of cellulitis by their primary care physician to continue to be managed by them (controls [n = 9]) or to receive a dermatology consultation (treatment group [n = 20]). Only 2 (10%) patients randomized to consultation had their diagnosis of cellulitis confirmed, as did only 3 of 9 controls; all 9 controls were treated with antibiotics, whereas only the 2 confirmed cases in the consultation group were so treated. All 20 patients in the group randomized to undergo consultation reported improvement at follow-up 1 week later. The most common etiologies of pseudocellulitis were eczema (including atopic and contact dermatitis), stasis dermatitis, erythema migrans, and arthropod reaction.

The lack of a specific test for cellulitis and the large number of its mimics [3] presumably accounts for its overdiagnosis. Patients with cellulitis may lack fever or leukocytosis and the value of inflammatory biomarkers is, at best, limited. One study reported that procalcitonin determination had only a 51.8% negative predictive value [4]. Overdiagnosis leads to unnecessary hospitalizations and its attendant costs, as well as the unnecessary administration of antibiotics. We need to do a better job.

References

Bacterial Evolution Continues, Even in the Absence of Environmental Influences

Lenski RE, Wiser MJ, Ribeck N, et al. Sustained fitness gains and variability in fitness trajectories in the long-term evolution experiment with Escherichia coli. Proc Biol Sci 2015;282. doi:10.1098/rspb.2015.2292.

Natural selection refers to the enrichment, within a population, of individuals with improved survival and reproductive fitness, thus allowing their favorable adaptation to their environment as well as species propagation. It is often assumed that, in the absence of changes in that environment, there is a limit to this process and that organisms will eventually reach a fitness peak, obviating further change. In such an unchanging environment, evolution would be best described as a hyperbolic model with an asymptotic upper limit. Lenski and colleagues provide data suggesting that a power-law model, in which one quantity continues to vary as a power of another, better describes evolution in a static environment. Although the power-law model also predicts a declining rate of improvement in fitness, there is no asymptote—that is, fitness improvement continues at a diminishing rate but without an upper limit.

Michigan State University's Long-Term Evolution Experiment dates to 1988, at which time a clonal strain of Escherichia coli was distributed into 12 portions with each incubated under identical conditions. They have repeatedly examined the fitness of 9 of the 12 relative to that of a reference strain, after 40 000, 50 000, and 60 000 generations in a growth competition assay using mixed cultures.

The investigators performed a total of 1100 such assays, finding that fitness continued to significantly increase over these generations, but at a decelerating (although not statistically significant) rate when compared with previous generations. There was lack of parallelism in evolutionary changes, as indicated by subtle divergences between the individual populations that affected their fitness trajectories. Thus, the “results imply that both adaptation and divergence can continue indefinitely—or at least for a long time—even in a constant environment”.

These results demonstrate the power of evolution, which appears to not require a changing environment to continue, albeit doing so at a slower rate than would occur in the real world, where organisms face a continually changing environment.